Use when optimizing codebase performance, reducing bundle sizes, improving build times, or eliminating code bloat. Launches parallel optimization hunters with mandatory before/after verification.
From speedrunnpx claudepluginhub wilrf/wilrf-marketplace --plugin speedrunThis skill uses the workspace's default tool permissions.
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.
Orchestrates subagents to execute phased plans: deploys for implementation, verification, anti-pattern checks, code quality review, and commits only after passing checks.
Comprehensive optimization hunting that dispatches parallel agents to find performance bottlenecks across all layers, then applies fixes with mandatory verification.
NO OPTIMIZATION CLAIM WITHOUT MEASUREMENT.
Every optimization must have before/after metrics. Unverified "improvements" are worthless.
All hunters use the same priority classification:
| Priority | Impact | Fix Effort |
|---|---|---|
| P0 | >50% improvement or >100KB savings | Trivial (1-2 lines) |
| P1 | 20-50% improvement or 50-100KB savings | Straightforward |
| P2 | 10-20% improvement or 10-50KB savings | Moderate |
| P3 | <10% improvement or <10KB savings | Any |
Confidence:
Fix order: P0 first. Never spend time on P3 while P0 exists.
| Hunter | Profile | Focus | Color |
|---|---|---|---|
| bundle-hunter | OBSESSIVE | File sizes, tree-shaking, code splitting | yellow |
| complexity-hunter | UNCOMFORTABLE | Cyclomatic complexity, nesting, function length | purple |
| dead-code-hunter | RUTHLESS | Unused exports, orphan files, unreachable paths | gray |
| dependency-hunter | WARY | Bloat, duplicates, outdated, vulnerabilities | magenta |
| build-hunter | IMPATIENT | Compilation time, caching, parallelization | orange |
| algorithm-hunter | IMPATIENT | Big O, data structures, repeated computation | red |
| query-hunter | METICULOUS | N+1, missing indexes, slow queries | green |
| web-vitals-hunter | OBSESSIVE | LCP, CLS, INP, TTFB | blue |
| memory-hunter | PARANOID | Leaks, allocation patterns, GC pressure | red |
| image-hunter | METICULOUS | Formats, compression, lazy loading | cyan |
Always dispatch: bundle, complexity, dead-code, dependency, algorithm
Dispatch if applicable:
build-hunter — has build config (webpack/vite/rollup/turbo)query-hunter — has database (Supabase/Prisma/SQL/Mongoose)web-vitals-hunter — has frontend (React/Vue/HTML)memory-hunter — has server runtime (Node/Python/Go)image-hunter — has image assets (public/, static/, assets/)# Speedrun Optimization Report
**Date:** YYYY-MM-DD
**Stack:** [detected tech stack]
**Baseline:** Bundle: X MB | Build: Xs | Complexity avg: Y
## Summary
| Hunter | P0 | P1 | P2 | P3 | Total |
|--------|----|----|----|----|-------|
| bundle-hunter | N | N | N | N | N |
| query-hunter | N | N | N | N | N |
| [etc.] | | | | | |
## P0 — Act Now
[Quick wins: high impact, trivial effort]
### [Issue Title]
- **Priority:** P0
- **Hunter:** bundle-hunter
- **Impact:** ~80KB savings
- **Confidence:** HIGH
- **Fix:** [What to do]
## P1 — High Value
[Straightforward fixes worth doing this sprint]
## P2 — Medium Value
[Moderate effort, good returns]
## P3 — Consider Later
[Minor gains, low urgency]
## Verification
| Metric | Before | After | Delta |
|--------|--------|-------|-------|
| Bundle | X MB | Y MB | -Z% |
| Build | Xs | Ys | -Z% |
| Queries/request | N | M | -Z% |
Re-measure all baseline metrics and report the delta. Refuse to claim an improvement without measurement.